The plant cell culture expression system has several advantages over the bacterial, yeast or Baculovirus expression systems. Bacteria do not, and yeasts only limitedly, carry out post-translational modifications of the expressed proteins. Plant cells are eukaryotic and able to perform sophisticated protein modifications which are often necessary for the proper function of proteins.
Although Baculovirus is a potent transformation vehicle for higher eukaryotes and generally performs satisfactory modifications of proteins, the cost for culturing baculovirus is much higher than that for plant cells. In addition, the host Cells are eventually lysed by Baculovirus and thousands of host proteins along with the expressed transformation protein are mixed and released into the culture medium, which makes purification of the expressed transformation protein difficult.
The culture medium for plant cells contains mainly salts and vitamins and therefore, it costs much less than that used to culture insect cell lines which are used for the Baculovirus transfection. Moreover, the culture medium for plant cells will not need a supply of serum, whereas almost all animal cell cultures cannot survive without serum. In addition, since plant cells are eukaryotes, the expressed proteins therein will be appropriately post-translationally modified so as to render said proteins capable of functioning and being secreted out of the plant cells. Although no one has yet made a deeper understanding of the mechanism of protein secretion in plant cells, the common belief at present is that it could be similar to the secretory mechanism in animals.
Plant cell cultures are a potential commercial source of medicines, dyes, enzymes, flavoring agents and aromatic oils. Plant cell culture production of such compounds are sought when (1) they are produced by the plant in small quantities or in fleeting or unharvestable developmental stages of the plant's life cycle; (2) when they are produced by plants which are not amenable to agriculture or are native to vanishing or inaccessible environments; and (3) when the compounds cannot be satisfactorily synthesized in vitro or by other biosynthesis systems.
Attempts to produce products by plant cell culture, however, are often commercially unsuccessful due to such factors as insufficient production and secretion of the desired product, poor cell growth, and difficulties in maintaining the appropriate cell type in culture.
The callus alpha-amylase (.alpha.-amylase) expression system has features which make it of potential use to plant cell fermentation technology, namely its high level of expression, sustained expression, expression irrespective of either the tissue of origin of the cell culture or tissue formation in the cell culture, and its product secretion. Although rice callus itself may not be an ideal source of commercial .alpha.-amylase, the gene regulatory regions responsible for the high expression could be used, with the aid of recombinant DNA technology and plant transformation, to achieve high expression of other valuable proteins (Carl R. Simmons, et al (1991), Biotechnology and Bioengineering, 38: 545-551).
Starch includes straight-chain starch and branched starch, two types of polysacchardies, and is the basic stored nutrient component in cereal grains (T. Akazawa et al (1985), Ann. Rev. Plant Physiol., 36: 441-472). During the initial germinating period of cereal seeds, the aleurone layer cells will synthesize .alpha.-amylase. Alpha-amylase, .alpha.-glucosidase and enzymes restricting dextrinase are secreted into the endosperm and together hydrolyze starch to form glucose and maltose, so as to provide the nutrients needed for the growth of the germ (J. C. Rogers and C. Milliman, J. Biol. Chem., 259 (19): 12234-12240, 1984; Rogers, J. C., J. Biol. Chem., 260: 3731-3738, 1985). Other enzymes contributing to starch hydrolysis include .beta.-amylase which can hydrolyze starch to form maltose and a small amount of glucose. In a dry seed, .beta.-amylase normally exists in an inactive form in the endosperm due to protein disulfide bonding. When the seed germinates, the aleurone layer cells will be subjected to the induction by gibberellic acid (GA.sub.3) to produce protease, which can destroy the disulfide bond and release the active form of .beta.-amylase. The above four enzymes take part in the hydrolysis of starch during the germination of seeds. However, .alpha.-amylase is the most active and holds the most important role (Akazawa, T., et al (1985), Ann. Rev. Plant Physiol., 36: 441-472).
It is known that GA.sub.3 exerts a direct influence over the expression of .alpha.-amylase (Chandler, P. M., et al (1984), Mol. Biol., 3: 401-418). When rice seeds are treated with GA.sub.3, the new synthesis of .alpha.-amylase mRNA by the aleurone layer cells increases to 50 to 100-fold of the control value (no GA.sub.3) (O'Neill, S. D., et al (1990), Mol. Gene. Genet., 221: 235-244). In reality, the regulation of .alpha.-amylase gene expression by GA.sub.3 has provided a very ideal model for studying the mechanism of hormonal regulation of gene expression in plants (Ho, T. H. D., et al (1987), .-+.Regulation of gene expression in barley aleurone layers," In: Molecular Biology of Plant Growth Control, pp. 35-49. St. Louis, Mo.: Alan R. Liss, Inc.).
Hitherto, .alpha.-amylase genes from rice, barley and wheat have been cloned and subjected to further study and analysis. The results show that these cereal-type .alpha.-amylase isozymes or isoforms are all manufactured by several types of .alpha.-amylase genes (Baulcombe, D. C., et al (1987) Mol., Gen. Genet., 209: 33-40); Huang, N., et al (1990a), Plant Mol. Biol., 14: 655-668; Knox, C. A. P., et al (1987), Plant Mol. Biol., 9: 3-17).
The .alpha.-amylase secreted from the aleurone layer cells during the germinating period of the seed of barley and wheat comprises two classes, the high isoelectric point and low isoelectric point. In barley, there are around 7 .alpha.-amylase genes which belong to the high isoelectric point and 3-4 genes which belong to the low isoelectric point (B. Khursheed and J. C. Rogers, J. Biol. Chem., 263: 18593-18960, 1988).
Currently, 7 .alpha.-amylase cDNA and 9 .alpha.-amylase genomic DNA groups of barley have been cloned (Chandler, P. M., et al (1984), Plant. Mol. Biol., 3: 401-418; J. Deikman and R. L. Jones, Plant Physiol., 78: 192-198, 1985; Khrusheed & Rogers (1988), supra; Knox, C. A. P., et al (1987), supra). The .alpha.-amylase genes of wheat are grouped into .alpha.-Amy1, .alpha.-Amy2 and .alpha.-Amy3. Alpha-Amy1 has a high isoelectric point while .alpha.-Amy2 has a low isoelectric point, and each has more than 10 genes which are expressed in germinating seeds. Alpha-amylase .alpha.-Amy3 includes 3-4 genes which are expressed in immature seeds (Baulcombe et al (1987), supra).
With regard to the study of rice .alpha.-amylase genes, the .alpha.-amylase genes thereof have not been classified into the high isoelectric point group and the low isoelectric point group as was done in the study of barleys and wheats. In reality, MacGregor, A. W., et al (Cereal Chem., 65: 326, 1988) applied the analytical method of isoelectric point electrophoresis and found that rice .alpha.-amylase isomers had a pI value of less than 5.5.
Therefore, it is possible that rice does not have any isoform of high isoelectric point. Huang, N., et al (Nucl. Acids. Res., 18: 7007-7014, 1990b) grouped the 10 rice .alpha.-amylase genes into 5 groups by cross hybridization experiment and confirmed their distribution in 5 chromosomes (Ranjhan et al, the original manuscript is still under preparation). O'Neill et al (Mol. Gene. Genet. 221: 235-244, 1990) made the first more detailed study of the cDNA pOS103 and pOS137 of rice .alpha.-amylase. The .alpha.-amylase manufactured from pOS103 and pOS137 has a precursor protein of a molecular weight of 48 KDa.
When this enzyme is secreted out of the cell, the signal peptide chain of the precursor protein will be cleaved off. Accordingly, the molecular weight of mature .alpha.-amylase is about 45-46 KDa and the isoelectric point thereof is predicted to be about 6.0. However, Kumagai, M. H., et al (Gene, 94: 209-216, 1990) subcloned pOS103 into the cells of Saccharomyces, to allow the Saccharomyces to secrete .alpha.-amylase into the culture medium, and it was found that the molecular weight of .alpha.-amylase is about 44-45 KDa and that the isoelectric point is about 4.7 to 5.0.
On the other hand, transformation of dicotyledonous plants with Agrobacterium tumefaciens is well established and widely used. A number of foreign genes carried between the T-DNA borders of the Ti plasmid in Agrobacterium have been delivered to plant cells, integrated into the chromosome, and stably inherited by subsequent generations. This, however, has not been the case for monocotyledonous plants in general. In the past, the monocots and particularly the graminaceous crop species have been considered to be outside the Agrobacterium host range (Bevan, M. W., Nucl. Acids Res., 12: 8711-8721, 1984; Declene, M., Phytopathol. Z. 113: 81-89). Gene transfer methods developed from economically important monocotyledonous species have been restricted to the directed transfer of DNA into protoplasts, or particle discharge methods of direct DNA transfer into intact cells of embryomic callus or suspension cells.
In recent years, more and more data on the transformation of monocots using Agrobacterium have been accumulated. The demonstration of Agrobacterium T-DNA integration into genomic DNA of Asparagus officinalis (Bytebier., B., et al (1987), Proc. Natl. Acad. Sci. USA, 84:5345-5349) and Dioscorea bulbifera (Schafer, W., et al (1987), Nature, 327: 529-531) first indicated that some monocot species possess the potential to be transformed by Agrobacterium. Later, a report of T-DNA integration into the genomic DNA of rice, Oryzae sativa (Raineri, D. M., et al (1990), Biotechnology, 8: 33-38), further showed that graminaceous crop plants can be transformed by Agrobacterium. Recently, foreign genes have been successfully transferred into corn, and regeneration of plants and detection of the transferred genes in the F1 progeny have been demonstrated (Gould, J. et al (1991), Plant Physiol., 95: 426-434). Therefore, the Agrobacterium-mediated gene transfer system seems to be applicable for transformation of monocot plants.
Agrobacterium-mediated transformation is a complex process and several factors are involved (for review, see Hooykaas, P. J. J., Plant Mol. Biol., 13: 327-336, 1989). Activation of the virulence system is one of the early important steps in plant tumor induction (Garfinkrl, D. J., J. Bacteriol., 144: 732-743, 1980). The vir genes on the Ti plasmid are silent until they become induced by certain plant factors, which in tobacco have been identified as the phenolic compounds acetosyringone and .alpha.-hydroxy-acetosyringone (Stachel, S. E., et al (1985), Nature, 318: 624-629). These compounds are released from plant tissue, especially after wounding, which has long been known to be a prerequisite for plant tumorigenesis via Agrobacterium. Although initially, it was generally thought that monocot species were not susceptible to Agrobacterium, some monocot species (e.g., Asparagus) are prone to tumor formation after T-DNA transfer (Hernalsteens, J. P., et al (1984), EMBO J., 3: 3039-3041). Tumor formation on discs of the monocot Dioscorea (yam) by Agrobacterium requires a pre-incubation with exudates from dicot plants (Schafer, W., et al (1987), Nature, 327: 529-531), indicating that some monocots probably do not produce enough inducers to activate the expression of the vir gene on the Ti plasmid transferred by Agrobacterium.
Toxins or inhibitors which inhibit the growth of Agrobacterium tumefaciens and the expresion of vir genes on the Ti plasmid have been shown to be present in wheat (Usami, S., et al (1988), Proc. Natl. Acad. Sci. USA, 85: 3748-3752), and corn (Sahi, S. U., et al (1991), Proc. Natl. Acad. Sci. USA, 87: 3879-3883), and might cause problems during attempts to transform monocots with Agrobacterium. Nevertheless, wheat and oats have been shown to contain substances which induce the expression of the vir locus of the Ti plasmid and the T-DNA processing reaction, although the inducing substance of wheat differs from acetosyringone (Usami, S., et al (1988), supra).
Previously, it was reported that potato suspension culture (PSC) is essential for the Agiobacterium-mediated transformation of Indica type rice (Chan, M. T., et al, "Transformation of Indica rice (Oryza sativa L.) mediated by Agrobacterium," Plant Cell Physiol. (1992), 33: 577-583). PSC is rich in the phenolic compounds acetosyringone (AS) and sinapinic acid (SA). Although the role of these two compounds in the success or efficiency of transformation is not yet known, the results imply that transformation of monocots, at least rice, using Agrobacterium can be improved by the addition of certain substances.
The age and physiological states of plant tissues have been shown to be important for Agrobacterium-mediated transformation (An, G. et al (1986), Plant Physiol., 81: 301-305; Chan, M. T., et al (1992), supra); H. H. Chang and M. T. Chan, Bot. Bull. Academia Sinica, 32: 171-178, 1991; Dale, P. J., et al (1989), Plant Sci., 63: 237; Gould, J. et al (1991), supra; Hernalsteens J. P., et al (1984), supra).
Thes studies suggest that infection with Agrobacterium and T-DNA transfer should take place in monocots if suitable tissues are used for transformation. It was previously shown that young tissues of rice root have a greater potential to be transformed by Agrobacterium if appropriate conditions are applied (Chan, M. T., et al (1992), supra), and it was assumed that young tissues may contain relatively fewer inhibitors or more virulence inducers. Therefore, a combination of immuture, embryos and PSC for transformation of rice can be used in the present invention.
This invention is based on the inventors' discovery that, in addition to regulation by gibberellic acid (GA.sub.3) in germinating seeds of rice, the expression of .alpha.-amylase genes in suspension-cultured cells of rice is regulated by the level of carbohydrate present in the culture medium (Yu, Su-May et al. (1991), J. Biol. Chem., 266: 21131-21137).
The synthesis of .alpha.-amylases and levels of their mRNA are greatly induced under sucrose starvation. An increase of .alpha.-amylase synthesis is assumed to accelerate hydrolysis of cellular starch as an energy source when exogenous carbon source is depleted. Under normal growth condition with an adequate supply of sugars in the medium, the expression of .alpha.-amylase genes is subject to metabolite repression. It was further observed that .alpha.-amylases synthesized by the cultured rice cells are secreted into the culture medium and can account for about 15-20% of the total proteins present in the medium during periods of sugar depletion.
It would therefore be advantageous to develop a gene expression system in plant cell culture by constructing a vector expressible in plant host cells utilizing the. promoter and the signal peptide sequences of an .alpha.-amylase gene. Any foreign gene can be linked downstream of said promoter and signal peptide encoding sequences. This construct would then be used to transform a compatible plant host cell.
Theoretically, the .alpha.-amylase promoter would control the expression of foreign genes in said plant cells and the secretion of the proteins into the medium. Such an expression system therefore has a high potential to express and/or secrete large quantities of any important protein into the medium, greatly facilitating purification of the expressed protein.
To aid in the procedure of screening and/or to enhance further the expression efficiency of the gene expression system constructed above, said gene expression system may further comprise a suitable marker gene, a reporter gene, an antibiotic-resistance gene and/or an enhancer gene, all of which can be those well known by an artisan of ordinary skill in the relevant art (Maniatis, T., et al, "Molecular Cloning: A Laboratory Mannual," pressed by Cold Spring Harbor Laboratory, 2nd edi., 1989).